Philip Hyunsu Cho 91537e7353

Fix #3342 and h2oai/h2o4gpu#625: Save predictor parameters in model file (#3856) ...

* Fix #3342 and h2oai/h2o4gpu#625: Save predictor parameters in model file

This allows pickled models to retain predictor attributes, such as
'predictor' (whether to use CPU or GPU) and 'n_gpu' (number of GPUs
to use). Related: h2oai/h2o4gpu#625

Closes #3342.

TODO. Write a test.

* Fix lint

* Do not load GPU predictor into CPU-only XGBoost

* Add a test for pickling GPU predictors

* Make sample data big enough to pass multi GPU test

* Update test_gpu_predictor.cu

2018-11-03 21:45:38 -07:00

..

base.h

Use __CUDA__ macro with __NVCC__. (#3539)

2018-08-02 22:04:23 +12:00

c_api.h

Add callback interface to re-direct console output (#3438)

2018-07-05 11:32:30 -07:00

data.h

Add basic unittests for gpu-hist method. (#3785)

2018-10-15 15:47:00 +13:00

feature_map.h

Clang-tidy static analysis (#3222)

2018-04-19 18:57:13 +12:00

gbm.h

Revert "Fix #3485, #3540: Don't use dropout for predicting test sets" (#3563)

2018-08-08 09:48:55 -07:00

learner.h

Fix #3342 and h2oai/h2o4gpu#625: Save predictor parameters in model file (#3856)

2018-11-03 21:45:38 -07:00

linear_updater.h

Implement GPU accelerated coordinate descent algorithm (#3178)

2018-04-20 14:56:35 +12:00

logging.h

Fix CRAN check by removing reference to std::cerr (#3660)

2018-09-05 11:44:00 -07:00

metric.h

Clang-tidy static analysis (#3222)

2018-04-19 18:57:13 +12:00

objective.h

Replaced std::vector with HostDeviceVector in MetaInfo and SparsePage. (#3446)

2018-08-30 14:28:47 +12:00

predictor.h

Dmatrix refactor stage 1 (#3301)

2018-06-07 10:25:58 +12:00

tree_model.h

Span class. (#3548)

2018-08-14 17:58:11 +12:00

tree_updater.h

Clang-tidy static analysis (#3222)

2018-04-19 18:57:13 +12:00